Upload
lenhi
View
262
Download
10
Embed Size (px)
Citation preview
Introduction & Basic Concepts of Thermodynamics
Reading Problems2-1→ 2-8 2-53, 2-67, 2-85, 2-96
Introduction to Thermal SciencesT
herm
od
yn
am
ics
Heat T
ran
sfer
Fluids Mechanics
Thermal
Systems
Engineering
Thermodynamics
Fluid Mechanics
Heat TransferConservation of mass
Conservation of energy
Second law of thermodynamics
Properties
Fluid statics
Conservation of momentum
Mechanical energy equation
Modeling
Conduction
Convection
Radiation
Conjugate
Thermodynamics: the study of energy, energy conversion and its relation to matter. The analysisof thermal systems is achieved through the application of the governing conservation equa-tions, namely Conservation of Mass, Conservation of Energy (1st law of thermodynamics),the 2nd law of thermodynamics and the property relations.
Heat Transfer: the study of energy in transit including the relationship between energy, matter,space and time. The three principal modes of heat transfer examined are conduction, con-vection and radiation, where all three modes are affected by thermophysical properties, geo-metrical constraints and the temperatures associated with the heat sources and sinks used todrive heat transfer.
Fluid Mechanics: the study of fluids at rest or in motion. While this course will not deal exten-sively with fluid mechanics we will be influenced by the governing equations for fluid flow,namely Conservation of Momentum and Conservation of Mass.
1
Thermodynamic Systems
Isolated Boundary
System
Surroundings
Work
Heat
System Boundary
(real or imaginary
fixed or deformable)
System
- may be as simple as a melting ice cube
- or as complex as a nuclear power plant
Surroundings
- everything that interacts
with the system
System: thermodynamic systems are classified as either closed or open
Closed System:
• consists of a fixed mass
• NO mass crosses the boundary
Work
Heat
Mass is fixed
Open System: Steady
• consists of a fixed volume
• mass crosses the boundary
• no change with respect to timeHeat
Work
Mass out
Mass in
Open System: Unsteady
• changes may occur with respect to time
• flow work in does not equal flow work out
• energy crosses the boundary as enthalpyand heat
Heat
Mass in
Flow work in
Mass changes
with time
2
Thermodynamic Properties of Systems
Thermodynamic Property: Any observable or measurable characteristic of a system or anymathematical combination of measurable characteristics
Extensive Properties: Properties that are dependent of the size or extent of the system, i.e. mass
• they are additive ⇒ XA+B = XA +XB
Intensive Properties: Properties that are independent of the size (or mass) of the system
• they are not additive ⇒ XA+B 6= XA +XB
Specific Properties: Extensive properties expressed per unit mass to make them intensiveproperties
• specific property (intensive) −→extensive property
mass
Measurable Properties
• P, V, T, andm are important because they are measurable quantities
– pressure (P ) and temperature (T ) are easily measured intensive properties.Note: They are not always independent of one another.
– volume (V ) and mass (m) are easily measured extensive properties
Pressure
• Pressure =Force
Area
• gauge = absolute - atmospheric⇒ Pabs > Patm
Pgauge = Pabs − Patm
• vacuum pressure⇒ Pabs < Patm
Pvac = Patm − Pabs
• thermodynamics properties depend onabsolute pressure
absolute
pressure
gauge
pressure
vacuum
pressure
absolute
vacuum pressure
ABSOLUTE
ATMOSPHERIC
PRESSURE
Pre
ssu
re
Patm
3
Temperature• temperature is a pointer for the direction of energy transfer as heat
Q QT
A
TA
TA
TA
TB
TB
TB
TB
> <
0th Law of Thermodynamics: when two objects are in thermal equilibrium with a third objectthey are in thermal equilibrium with each other.
• the 0th law makes a thermometer possible
CC
A
T = TA C
B
T = TB C
T = TA B
• in accordance with the zeroth law, any system that possesses an equation of state that relatesT to other accurately measurable properties can be used as a thermometer
Other Properties• energy within a system can be stored as a combination of kinetic energy, potential energy or
internal energy
• Internal energy U [kJ ]:
– associated with molecular motion i.e. translational, rotational or vibrational
– extensive since it depends on the amount of matter in the system
• Kinetic energy KE:
– the energy of motion relative to some reference frame ⇒ KE =1
2m(V)2
• Potential energy PE:
– the energy of position within a gravitational field ⇒ PE = mgz
4
State and Equilibrium
• the state of a system is its condition as described by a set of relevant energy related properties.
• a system at equilibrium is in a state of balance
DefinitionsSimple Compressible System: A simple compressible system experiences negligible electrical,
magnetic, gravitational, motion, and surface tension effects, and only PdV work is done.
State Postulate:
State Postulate (for a simple compressible system): The state of a simplecompressible system is completely specified by 2 independent and intensive properties.
Processes and CyclesDefinitions
1. Process: a transformation from one equilibrium state to another through a change in prop-erties
2. Quasi-equilibrium Process: changes occurs sufficiently slow to allow the system to tran-sition in a uniform manner
3. Cycle: a sequence of processes that begin and end at the same state i.e. see the Carnot cycle
4. Steady: no change with respect to time
• if the process is not steady, it is unsteady or transient
• often steady flow implies both steady flow and steady state
5. Uniform: no change with respect to position
• if the flow field in a process is not uniform, it is distributed.
5
Example 1-1: A glass tube is attached to a water pipe. If the water pressure at the bottomof the glass tube is 115 kPa and the local atmospheric pressure is 92 kPa, determine howhigh the water will rise in the tube, in m. Assume g = 9.8 m/s2 at that location and takethe density of water to be 1000 kg/m3.
Step 1: Draw a clearly labeled diagram to represent the system
Step 2: State what the problem is asking you to determine.
Step 3: State all assumptions used during the solution process.
Step 4: Prepare a table of properties.
Step 5: Solve (start by writing a force balance on the system)
Step 6: Clearly identify your answer
6
Properties of Pure Substances
Reading Problems4-1→ 4-7 4-41, 4-54, 4-58, 4-79, 4-955-3→ 5-5
Pure Substances• it has a fixed chemical composition throughout (chemically uniform)
• a homogeneous mixture of various chemical elements or compounds can also be consideredas a pure substance (uniform chemical composition)
Phases of Pure SubstancesSolids: strong molecular bonds
Liquids: molecules are no longer in a fixed position relative to one another
Gases: there is no molecular order
Behavior of Pure Substances (Phase Change Processes)• all pure substances exhibit the same
general behavior.
– one exception is that most con-tract on freezing. H2O (and afew more) expand on freezing.
• phase change processes:
• Critical point: is the point at whichthe liquid and vapor phases are not dis-tinguishable (we will talk about thismore)
• Triple point: is the point at which theliquid, solid, and vapor phases can existtogether
sublimatio
n
vaporizatio
n
melti
ng
meltin
g
triple point
critical point
SOLID
VAPOR
LIQUID
vaporization
condensation
sublimation
sublimation
freezingmelting
P
T
substances that
expand on freezing
substances that
contract on freezing
Phase Diagram
1
Critical Point1 Triple Point2
P (MPa) T (◦C) P (MPa) T (◦C)
H2O 22.06 373.95 0.00061 0.01
O2 5.08 −118.35 0.000152 −218.79
CO2 7.39 31.05 0.517 56.6
N 3.39 −146.95 0.0126 −209.97
He 0.23 −267.85 0.0051 −270.96
1 - see Table A-1 for other substances 2 - see Table 4.3 for other substances
T − v Diagram for a Simple Compressible Substance
consider an experiment in which a substance starts as a solid and is heated up at constant pressureuntil it all becomes as gas
P0
P0
P0
P0
P0
P0
dQ dQ dQ dQ dQ
S S
L LL
G G
S + G
L + G
SS + L
G
L
T
v
critical point
sta urate
dvap
orline
triple point line
satu
rate
dliq
uid
line
fusio
nlin
e
2
The P-v-T Surface Diagram
• if P-v-T is projected onto a pressure-temperature plane we obtain a phase diagram
• if we project onto a pressure-volume plane we obtain the familiar P-v diagram for a substance(likewise for T-v)
The Vapor Dome (Two Phase Region)
v
critical point
vfg
subcooled
liquid
saturated
liquid line
L+G
vfvg
P
P (T)sat
saturated
vapor line
gas (vapor)
two-phase region
(saturated liquid &
saturated vapor)
Tcr
T
T
specific volume:vf , vg and vfg = vg − vf
internal energy:uf , ug and ufg = ug − uf
specific enthalpy:hf , hg and hfg = hg − hf
specific entropy:sf , sg and sfg = sg − sf
• the subscripts denote: f - saturated liquid (fluid) and g - saturated vapor (gas)
• given the shape of the vapor dome, as T ↑ or P ↑ ⇒ hfg ↓
3
Quality
We need to define a new thermodynamic property: Quality≡ x =mg
m=
mg
mg +mf
L
Gx = m /mg
1-x = m /mf
total mass of mixture ⇒ m = mg +mf
mass fraction of the gas ⇒ x = mg/m
mass fraction of the liquid ⇒ 1− x = mf/m
Properties of Saturated Mixtures
• To fix the state we require(T, x), (T, v), (P, x) . . .
• all the calculations done in the vapor domecan be performed using Tables.
– in Table A-4, the properties are listedunder Temperature
– in Table A-5, the properties are listedunder Pressure
• other properties are calculated as mass-average values of the saturated quantities
v
critical point
vfg
saturated liquid line: x=0
L
G
vfv
mixture
vg
P
P (T)sat
saturated vapor line: x=1
T
T
v =V
m=Vf + Vg
m=mfvf +mgvg
m= (1− x) vf + x vg
= vf + x(vg − vf) = vf + x vfg
x =v − vfvfg
u = (1− x)uf + x ug = uf + x ufg ⇒ x =u− ufufg
h = (1− x)hf + x hg = hf + x hfg ⇒ x =h− hfhfg
s = (1− x)sf + x sg = sf + x sfg ⇒ x =s− sfsfg
4
Properties of Superheated Vapor• superheated means T > Tsat at the prevailing P , eg. water at 100 kPa has a saturation
temperature of Tsat(P ) = 99.63 ◦C.
• To fix the state we require (T, v), (P, T ) . . .
v
L
L+GG
T
T (P)sat
P
P
P = 100 kPa
state: 200 C and 100 kPao
T = 200 Co
• superheated vapor is a single phase
– T and P are independent of eachother in the single-phase region
– properties are typically calculated asa function of T and P
• if T >> Tcr orP << Pcr then ideal gasbehavior applies and the ideal gas equa-tions of state can be used
• Table A-6 for superheated water
Properties of Sub-cooled (Compressed) Liquid
• sub-cooled liquid means T < Tsat at the prevailing P , eg. water at 20 ◦C and 100 kPahas a saturation temperature of Tsat(P ) = 99.63 ◦C.
v
L L+GG
T
T (P)sat
P
P
P = 100 kPa
state: 20 C and 100 kPao
T = 20 Co
• can be treated as incompressible (indepen-dent of P )
v = v(T, P )⇒ v ≈ v(T ) = vf(T )
u = u(T, P )⇒ u ≈ u(T ) = uf(T )
s = s(T, P )⇒ s ≈ s(T ) = sf(T )
• for enthalpy:
h(T, P ) = u(T, P ) + v(T, P )P ≈ uf(T ) + vf(T )P
= uf(T ) + vf(T )Psat(T )︸ ︷︷ ︸hf (T )
+vf(T )[P − Psat(T )]
= hf(T ) + vf(T )[P − Psat(T )]
5
Example 2-1: A saturated mixture of R-134a with a liquid volume fraction of 10% and apressure of 200 kPa fills a rigid container whose volume is 0.5 m3. Find the quality of thesaturated mixture and the total mass of the R134-a.
Equations of State: Calculate the Properties of Gaseous Pure SubstancesTables: water (Tables A-4→ A-8, A-15), R134a (Tables A-11→ A-13, A-16)
Graphs: water (Figures A-9 & A-10), R134a (Figure A-14)
Equations: (Table A-2c), (Table 4-4)
• a better alternative for gases: use the Equation of State which is a functional relationshipbetween P, v, and T (3 measurable properties)
Ideal Gases
• gases that adhere to a pressure, temperature, volume relationship
Pv = RT or PV = mRT
are referred to as ideal gases
– whereR is the gas constant for the specified gas of interest (R = Ru/M )
Ru = Universal gas constant, ≡ 8.314 kJ/(kmol ·K)
M = molecular wieght (or molar mass) of the gas (see Table A-1))
Real Gases
• experience shows that real gases obey the following equation closely:
Pv = ZRT (T and P are in absolute terms)
where Z is the compressibility factor
• if we “reduce” the properties with respect to the values at the critical point, i.e.
reduced pressure = Pr =P
PcPc = critical pressure
reduced temperature = Tr =T
TcTc = critical temperature
6
Reference Values for u, h, s
• values of enthalpy, h and entropy, s listed in the tables are with respect to a datum where wearbitrarily assign the zero value. For instance:
Table A-4, A-5: saturated liquid - the reference for both hf and sf is taken at the triplepoint and Tref = 0.01 ◦C. This is shown as follows:
uf(@T = 0.01 ◦C) = 0 kJ/kg
hf(@T = 0.01 ◦C) = 0 kJ/kg
sf(@T = 0.01 ◦C) = 0 kJ/kg ·K
Table A-11, A-12, & A-13: saturated R134a - the reference for both hf and sf is taken as−40 ◦C. This is shown as follows:
hf(@T = −40 ◦C) = 0 kJ/kg
hf(@T = −40 ◦C) = 0 kJ/kg
sf(@T = −40 ◦C) = 0 kJ/kg ·K
Others: sometimes tables will use 0K as the reference for all tables. While this standard-izes the reference, it tends to lead to larger values of enthalpy and entropy.
• for the most part the choice of a reference temperature in the property tables does not affect usbecause we generally deal with ∆h,∆u, or ∆s. The calculated difference of properties willbe identical regardless of the reference temperature. Note: Be careful not to use values fromdifferent tables in the same calculation in the event that a difference reference temperature isused in the tables.
7
Calculation of the Stored Energy• how do we calculate ∆U and ∆H?
1. one can often find u1 and u2 in the thermodynamic tables (like those examined for thestates of water).
2. we can also explicitly relate ∆U and ∆H to ∆T (as mathematical expressions) byusing the thermodynamic properties Cp and Cv.
• the calculation of ∆u and ∆h for an ideal gas is given as
∆u = u2 − u1 =∫ 2
1Cv(T ) dT (kJ/kg)
∆h = h2 − h1 =∫ 2
1Cp(T ) dT (kJ/kg)
• to carry out the above integrations, we need to knowCv(T ) andCp(T ). These are availablefrom a variety of sources
Table A-2a: for various materials at a fixed temperature of T = 300K
Table A-2b: various gases over a range of temperatures 250K ≤ T ≤ 1000K
Table A-2c: various common gases in the form of a third order polynomial
Specific Heats: Ideal Gases• for an ideal gas Pv = RT
Cv =du
dT⇒ Cv = Cv(T ) only
Cp =dh
dT⇒ Cp = Cp(T ) only
h = u+ Pv ⇒ h = u+RT ⇒ RT = h(T )− u(T )
If we differentiate with respect to T
R =dh
dT−du
dT
R = Cp − Cv
8
Specific Heats: Solids and Liquids• solids and liquids are incompressible. (i.e., ρ = constant) and it can be shown (mathe-
matically) that:
Cp = Cv = C
Ideal gases: ∆u =∫ 2
1C(T ) dT ≈ Cavg ∆T
∆h = ∆u+ ∆(Pv) = ∆u+ v∆P + P∆v↗0
≈ Cavg∆T + v∆P
Solids: ∆h = ∆u = Cavg∆T (v∆P is negligible)
Liquids: If P = const, ∆h = Cavg∆T
If T = const, ∆h = v∆P
Example 2-2: Using water as the working fluid, complete the table shown below. Usinggraph paper, create a separate T − v diagram for each state point that clearly shows how youhave located the state point (approximately to scale). For instance, using State Point 1 as anexample, show a curve of T = 400 ◦C and a curve of P = 0.8 MPa. State Point 1 willbe at the intersection of these two curves. The intersection should have the correct relationshipto the vapor dome (i.e. inside or outside).
For each case, show all calculations necessary to establish the actual location of the statepoint. Identify the relevant charts or tables used to perform your calculations
In the table, under the column for quality (x), indicate the quality of the state point is underthe dome, CL if the state point is in the compressed liquid region and SV if the state pointis in the superheated vapor region.
State T P v x◦C MPa m3/kg
1 400 0.82 200 0.0053 3.5 0.054 0.01 20.05 0.2 0.306 100 5.07 300 0.58 70.0 0.609 250 0.125
9
First Law of Thermodynamics
Reading Problems3-2→ 3-7 3-40, 3-54, 3-1055-1→ 5-2 5-8, 5-25, 5-29, 5-37, 5-40, 5-42, 5-63, 5-74, 5-84, 5-1096-1→ 6-5 6-44, 6-51, 6-60, 6-80, 6-94, 6-124, 6-168, 6-173
Control Mass (Closed System)In this section we will examine the case of a control surface that is closed to mass flow, so that nomass can escape or enter the defined control region.
Conservation of MassConservation of Mass, which states that mass cannot be created or destroyed, is implicitly satisfiedby the definition of a control mass.
Conservation of EnergyThe first law of thermodynamics states “Energy cannot be created or destroyed it can only changeforms”.
energy entering - energy leaving = change of energy within the system
Sign ConventionCengel Approach
Heat Transfer: heat transfer to a system is positive and heat transfer from a system is negative.
Work Transfer: work done by a system is positive and work done on a system is negative.
For instance: moving boundary work is defined as:
Wb =∫ 2
1P dV
During a compression process, work is done on the systemand the change in volume goes negative, i.e. dV < 0. Inthis case the boundary work will also be negative.
1
Culham Approach
Using my sign convention, the boundary work is defined as:
Wb = −∫ 2
1P dV
During a compression process, the change in volumeis still negative but because of the negative sign on theright side of the boundary work equation, the bound-ary work directed into the system is considered positive.Any form of energy that adds to the system is considered positive.
Example: A Gas Compressor
Performing a 1st law energy balance:
InitialEnergyE1
+−
{Energy gainW1−2
Energy loss Q1−2
}=
FinalEnergyE2
E1 +W1−2 −Q1−2 = E2 ⇒ ∆E = Q−W
A first law balance for a control mass can also be written in differential form as follows:
dE = δQ− δW
Note: d or ∆ for a change in property and δ for a path function
2
The differential form of the energy balance can be written as a rate equation by dividing throughby dt, a differential time, and then letting dt→ 0 in the limit to give
dE
dt=δQ
dt−δW
dt⇒
dE
dt= Q− W
where
dE
dt= rate of energy increase within the system, ≡
dU
dt+d KE
dt+d PE
dt
Q = rate of heat transfer
W = rate of work done, ≡ power
• most closed systems encountered in practice are stationary i.e. the velocity and the elevationof the center of gravity of the system remain constant during the process
• for stationary systems we can assume thatd KE
dt= 0 and
d PE
dt= 0
Example 3-1: During steady-state operation, a gearbox receives Win = 60 kW throughthe input shaft and delivers power through the output shaft. For the gearbox as the system, therate of energy transfer by heat is given by Newton’s Law of Cooling as W = hA(Tb − Tf).where h, the heat transfer coefficient, is constant (h = 0.171 kW/m2 · K) and the outersurface area of the gearbox is A = 1.0 m2. The temperature of the outer surface of the gear-box is Tb = 300 K and the ambient temperature surrounding the gearbox is T∞ = 293 K.Evaluate the rate of heat transfer, Q and the power delivered through the output shaft, Wout.
aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
aaaaaaaaa
aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
aaaaaaaaa
aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
aaaaaaaaa
aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
aaaaaa
aaaaaa
aaaaaa
aaaaaa
aaaaaa
aaaaaa
aaaaaa
aaaaaa
aaaaaa
aaaaaa
aaaaaa
aaaaaa
aaaaaa
aaaaaa
aaaaaa
aaaaaa
aaaaaa
aaaaaa
W = 60 kWin
Wout
Q = hA(Tb-T )f
T = 293 Kf
T = 300 Kb
A = 1.0 m2
.
.
.
3
Forms of Energy Transfer
Work Versus Heat• Work is macroscopically organized energy transfer.
• Heat is microscopically disorganized energy transfer.
more on this when we discuss entropy
Heat Energy• heat is defined as a form of energy that is transferred solely due to a temperature difference
(without mass transfer)
• heat transfer is a directional (or vector) quantity with magnitude, direction and point of action
• modes of heat transfer:
– conduction: diffusion of heat in a stationary medium (Chapters 10, 11 & 12)
– convection: it is common to include convective heat transfer in traditional heat transferanalysis. However, it is considered mass transfer in thermodynamics. (Chapters 13 &14)
– radiation: heat transfer by photons or electromagnetic waves (Chapter 15)
Work Energy• work is a form of energy in transit. One should not attribute work to a system.
• work (like heat) is a “path function” (magnitude depends on the process path)
• work transfer mechanisms in general, are a force acting over a distance
Mechanical Work
W12 =∫ 2
1F ds
• if there is no driving or resisting force in the process (e.g. expansion in a vacuum) or theboundaries of the system do not move or deform,W12 = 0.
Moving Boundary Work
W12 = −∫ 2
1F ds = −
∫ 2
1P ·A ds = −
∫ 2
1P dV
• a decrease in the volume, dV → −ve results in work addition (+ve) on the system
4
• consider compression in a piston/cylinder,whereA is the piston cross sectional area (fric-tionless)
• the area under the process curve on a P − Vdiagram is proportional to
∫ 2
1P dV
• the work is:
– +ve for compression
– −ve for expansion
• sometimes called P dV work orcompression /expansion work
• polytropic processes: where PV n = C
• examples of polytropic processes include:
Isobaric process: if n = 0 then P = C and we have a constant pressure process
Isothermal process: if n = 1 then from the ideal gas equation PV = RT and PV isonly a function of temperature
Isometric process: if n → ∞ then P 1/nV = C1/n and we have a constant volumeprocess
Isentropic process: if n = k = Cp/Cv then we have an isentropic process. (tabulatedvalues for k are given in Table A-2) If we combine
Pvk = C with Pv = RT
we get the isentropic equations, given as:T2
T1
=
(v1
v2
)k−1
=
(P2
P1
)(k−1)/k
Case 1: for an ideal gas with n = 1 → W12 = −C lnV2
V1
Case 2: for n 6= 1 → W12 =P1V1 − P2V2
1− n(in general)
→ W12 =mR(T1 − T2)
1− n(ideal gas)
5
Example 3-2: A pneumatic lift as shown in the figure below undergoes a quassi-equilibriumprocess when the valve is opened and air travels from tank A to tank B.
A AB
Bair
atm.
valve closed valve open
control mass
system
Ap
mp
state 1 state 2
The initial conditions are given as follows and the final temperatures can be assumed to be thesame as the initial temperatures.
Patm = 100 kPa 1 Pa = 1 N/m2
mp = 500 kg Ap = 0.0245 m2
VA,1 = 0.4 m3 VB,1 = 0.1 m3
PA,1 = 500 kPa PB,1 = 100 kPaTA,1 = 298 K TB,1 = 298 K = 25 ◦C
Find the final pressures PA,2 and PB,2 and the work, W12, in going from state 1 to state 2.
Control Volume (Open System)The major difference between a Control Mass and and Control Volume is that mass crosses thesystem boundary of a control volume.
CONSERVATION OF MASS:
Unlike a control mass approach, the control volume approach does not implicitly satisfyconservation of mass, therefore we must make sure that mass is neither created nor destroyedin our process.
{rate of increase ofmass within the CV
}=
{net rate of
mass flow IN
}−{
net rate ofmass flow OUT
}
6
CONSERVATION OF ENERGY:
ECV (t) + δQ+ δWshaft + (∆EIN −∆EOUT )+
(δWIN − δWOUT ) = ECV (t+ ∆t) (1)
What is flow work?This is the work required to pass the flow across the system boundaries.
∆mIN = ρIN
volume︷ ︸︸ ︷AIN VIN ∆t
δWIN = F · distance
= PIN AIN︸ ︷︷ ︸F
· VIN ∆t︸ ︷︷ ︸∆s
=PIN ∆mIN
ρIN
since v = 1/ρ
7
δWIN = (P v ∆m)IN → flow work (2)
Similarly
δWOUT = (P v ∆m)OUT (3)
Substituting Eqs. 2 and 3 into Eq. 1 gives the 1st law for a control volume
ECV (t+ ∆t)− ECV (t) = δQ+ δWshaft + ∆mIN(e+ Pv)IN
−∆mOUT (e+ Pv)OUT (4)
Equation 4 can also be written as a rate equation→ divide through by ∆t and take the limitas ∆t→ 0
d
dtECV = Q+ Wshaft + [m(e+ Pv)]IN − [m(e+ Pv)]OUT
where:
e+ Pv = u+ Pv︸ ︷︷ ︸ +(V)2
2+ gz
= h(enthalpy) + KE + PE
Example 3-3: Determine the heat flow rate, Q, necessary to sustain a steady flow processwhere liquid water enters a boiler at 120 ◦C and 10 MPa and exits the boiler at 10 MPaand a quality of 1 for a mass flow rate is 1 kg/s. The effects of potential and kinetic energyare assumed to be negligible.
L
G
H O ( )2 l
in
out
steam
Q
8
Example 3-4: Steam with a mass flow rate of 1.5 kg/s enters a steady-flow turbine with aflow velocity of 50 m/s at 2 MPa and 350 ◦C and leaves at 0.1 MPa, a quality of 1,and a velocity of 200 m/s. The rate of heat loss from the uninsulated turbine is 8.5 kW .The inlet and exit to the turbine are positioned 6 m and 3 m above the reference position,respectively. Determine the power output from the turbine.Note: include the effects of kinetic and potential energy in the calculations.
Q
W
in
out
The Carnot CycleIf the heat engine is a reversible system where no entropy is generate internally, we refer to thecycle as the Carnot cycle.
T
T
T
s s s
Q
Q
Q
W
P = P
P = P
H
H
L
L
1
1
2
4
4
3
in
out
where the efficiency is given as:
η = 1−TL
TH⇐ Carnot efficiency
9
Practical Problems
• at state point 1 the steam is wet at TL and it is difficult to pump water/steam (two phase) tostate point 2
• can we devise a Carnot cycle to operate outside the wet vapor region
T
s s s
Q
Q
P = P
P
P
H
L
1
1
4
4
3
2
– between state points 2 and 3 the vapor must be isothermal and at different pressures -this is difficult to achieve
– the high temperature and pressure at 2 and 3 present metallurgical limitations
The Ideal Rankine Cycle
T
s
Q
Q
W
W
P = P
P = P
H
L
1
2
4
3
Q
Q
L
H
T
P
Boiler
Turbine
CondenserPump
• water is typically used as the working fluid because of its low cost and relatively large valueof enthalpy of vaporization
10
Device 1st Law Balance
Boiler h2 + qH = h3 ⇒ qH = h3 − h2 (in)
Turbine h3 = h4 + wT ⇒ wT = h3 − h4 (out)
Condenser h4 = h1 + qL ⇒ qL = h4 − h1 (out)
Pump h1 + wP = h2 ⇒ wP = h2 − h1 (in)
The net work output is given as
wT − wp = (h3 − h4)− (h2 − h1) = (h3 − h4) + (h1 − h2)
The Rankine efficiency is
ηR =net work output
heat supplied to the boiler=
(h3 − h4) + (h1 − h2)
(h3 − h2)
Example 3-5: For the steam power plant shown below,
T
s
Q
Q
W
W
P = P
P = P
H
L
1
2
4
3
Q
Q
L
H
T
P
Boiler
Turbine
CondenserPump
find: QH, QL, Wnet = WT − WP , and the overall cycle efficiency, ηR given the followingconditions:
P1 = 10 kPa P2 = 10 MPaP3 = 10 MPa P4 = 10 kPaT1 = 40 ◦C T3 = 530 ◦CT2 = 110 ◦Cx4 = 0.9 m = 5 kg/s
11
Entropy and the Second Law of Thermodynamics
Reading Problems7-1→ 7-3 7-88, 7-131, 7-1357-6→ 7-10 8-24, 8-44, 8-46, 8-60, 8-73, 8-99, 8-128, 8-132,8-1→ 8-10, 8-13 8-135, 8-148, 8-152, 8-166, 8-168, 8-189
The 1st law allows many processes to occur spontaneously in either direction, but our experiencetells us that they are not reversible.
Case 1: Hot object cooling down in atmosphere
Time
possible
not possible
atmospheric air at T0 atmospheric air at T0
+
object @
T > Ti 0
T < T < T0 iT0
Time
• object at Ti dissipates heat to the at-mosphere at T0 over time
• this process is not reversible i.e. en-ergy dissipated to the atmospherecannot be used to heat the objectback to Ti
Case 2: Pressurized cylinder releases air to the environment
possible
not possible
atmospheric air at P0
air at
P > Pi 0
air air @
P0
P < P < P0 i
atmospheric air at P0
+
• air held at Pi in a closed tank isspontaneously released to the sur-roundings
• this process is not reversible i.e.atmospheric air will not spon-taneously charge the tank eventhough energy is conserved
Case 3: A mass is allowed to fall from zi
possible
not possible
zi0 < z < zi
mass
mass
mass
• a mass suspended by a cable at ele-vation zi is released and allowed tofall
• this process is not reversible i.e.potential energy lost to internal en-ergy via heating and compressioncannot be used to raise the object
1
While the 1st law allowed us to determine the quantity of energy transfer in a process it doesnot provide any information about the direction of energy transfer nor the quality of the energytransferred in the process. In addition, we can not determine from the 1st law alone whether theprocess is possible or not. The second law will provide answers to these unanswered questions.
1. Direction:
Consider an isolated system whereQ = W = 0 (Assume energy to drive the propeller is atensioned spring, and no work crosses the boundary.)
gasgas
possible
impossible
State 1: cold system,
rotating propeller
& gas
State 2: warm system,
stationary propeller
& gas
2. Quality of Energy:A heat engine produces reversible work as it transfers heatfrom a high temperature reservoir at TH to a low temperaturereservoir at TL. If we fix the low temperature reservoir atTL = 300 K, we can determine the relationship betweenthe efficiency of the heat engine,
η = 1−QL
QH
= 1−TL
TH
as the temperature of the high temperature reservoir changes.In effect we are determining the quality of the energy trans-ferred at high temperature versus that transferred at low tem-perature.
TH
TL
reversible
heat engine Wrev
QH
QL
Second Law of Thermodynamics
DefinitionThe entropy of an isolated system can never decrease. When an isolated systemreaches equilibrium, its entropy attains the maximum value possible under theconstraints of the system
• we have conservation of mass and energy, but not entropy. Entropy is not conserved.
2
• the 2nd law dictates why processes occur in a specific direction i.e., Sgen cannot be−ve
• the second law states: (∆S)system + (∆S)surr. ≥ 0. If heat is leaving the system, then∆Ssystem can be negative but the combined entropy of the system and the surroundings cannever be negative.
System
Surroundings
Work
Heat
Non-Isolated Systems
- their entropy may decrease
Isolated System
- its entropy may
never decrease
Entropy
1. Like mass and energy, every system has entropy.
Entropy is a measure of the degree of microscopic disorder and represents our uncertaintyabout the microscopic state.
2. Reference: At T = 0K S = 0. There is no uncertainty about the microscopic state.
3. Relationship to Work: Microscopic disorder results in a loss of ability to do useful work.
4. Work: Energy transfer by work is microscopically organized and therefore entropy-free.
5. Heat: Energy transfer as heat takes place as work at the microscopic level but in a random,disorganized way. This results in entropy flow in or out of the system.
Reversible ProcessIn a reversible process things happen very slowly, without any resisting force, without any spacelimitation→ Everything happens in a highly organized way (it is not physically possible - it is anidealization).
An internally reversible system is one in which there are no irreversibilities in the system itselfbut there may be irreversibilities in the surroundings such as in the transfer of heat across a finitetemperature difference between the system and the surroundings.
3
Example: Slow adiabatic compression of a gas
→ idealization where S2 = S1 ⇒ Sgen = 0
T2 > T1 ⇒ increased microscopic disorder
V2 < V1 ⇒ reduced uncertainty about the whereabouts of molecules
Reversible︸ ︷︷ ︸Sgen=0
+ Adiabatic Process︸ ︷︷ ︸Q=0
⇒ Isentropic Process︸ ︷︷ ︸S1=S2
Does:
Isentropic Process⇒ Reversible + Adiabatic
NOT ALWAYS - the entropy increase of a substance during a process as a result of irreversibilitiesmay be offset by a decrease in entropy as a result of heat losses.
If Sgen =Q
Tsur, and
Sgen > 0 (fluid friction)
⇒ not reversible
Q > 0 ⇒ (non-adiabatic system)
If Sgen and Q/TTER and ∆S = 0, i.e. thesystem is isentropic, however, the system is notadiabatic nor is it reversible.
4
Evaluating EntropyFor a simple compressiblesystem, Gibb’s equations aregiven as:
Tds = du+ Pdv
Tds = dh− vdP
• these relations are valid for open and closed, and reversible and irreversible processes
T
s
dA = T ds
process path
1
2
ds
• the area under the curve on a T − s diagram isthe heat transfer for internally reversible pro-cesses
qint,rev =∫ 2
1T ds
qint,rev,isothermal = T∆s
• one can use Tds relations to calculate ds and∆s
Tabulated Calculation of ∆s for Pure SubstancesCalculation of the Properties of Wet Vapor:
Use Tables A-4 and A-5 to find sf , sg and/or sfg for the following
s = (1− x)sf + xsg s = sf + xsfg
Calculation of the Properties of Superheated Vapor:Given two properties or the state, such as temperature and pressure, use Table A-6.
Calculation of the Properties of a Compressed Liquid:Use Table A-7. In the absence of compressed liquid data for a property sT,P ≈ sf@T
Calculation of ∆s for Incompressible Materials
Tds = du+ Pdv for an incompressible substance, dv = 0, and Cp = Cv = C
ds =du
T= C
dT
T⇒ s2 − s1 =
∫ 2
1C(T )
dT
T
∆s = Cavg lnT2
T1
where Cavg = [C(T1) + C(T2)]/2
- if the process is isentropic, then T2 = T1, and ∆s = 0
5
Calculation of ∆s for Ideal Gases
Ideal Gas Equation ⇒ Pv = RT
du = Cv dT ⇒ u2 − u1 = Cv(T2 − T1)
dh = Cp dT ⇒ h2 − h1 = Cp(T2 − T1)
There are 3 forms of a change in entropy as a function of T & v, T & P , and P & v.
s2 − s1 = Cv lnT2
T1
+R lnv2
v1
= Cp lnT2
T1
−R lnP2
P1
= Cp lnv2
v1
+ Cv lnP2
P1
Variable Specific Heat
If we don’t assume Cv and Cp to be constant i.e. independent of T , then we can use the ideal gastables (Table A-21) as follows
s2 − s1 = so(T2)− so(T1) +R ln(v2/v1) or
s2 − s1 = so(T2)− so(T1)−R ln(P2/P1)
Non-Reversible Processes and Isentropic EfficienciesReal processes, such as expansion of steam in a turbine, produce irreversibilities associated withfriction, mixing, etc. The net result is that the performance is downgraded, resulting in a loss ofnet work output attributed to an increase in entropy.
For a turbine under steady flow conditions, the theoretical maximum work output assuming nointernal irreversibilities for an adiabatic process, i.e. isentropic expansion is
wisentropic = h1 − h2s
However for a real process where expansion leads to an increase in entropy, the actual work outputis degraded
wactual = h1 − h2
6
s
h
h -h1 2
P1
P2
h -h1 2s
1
2
2s
isentropic expansion
actual expansion
increase in entropy
Work Output Devices: i.e. Turbines and Nozzles
The ratio of the actual work to the isen-tropic work is called the isentropic effi-ciency and is defined as follows:
Turbine:
ηT =actual turbine work
isentropic turbine work
=h1 − h2
h1 − h2s
In a similar manner, the isentropic efficiency for other steady flow devices can be defined as fol-lows:
s
h
h -h2 1
P1
P2
h -h2s 1
1
2
2s
isentropic compression
actual compression
increase in entropy
Work Input Devices: i.e. Compressors and Pumps
Compressor:
ηC =Isentropic compressor work
actual compressor work
=h2s − h1
h2 − h1
Pump:
ηP =Isentropic pump work
actual pump work
=h2s − h1
h2 − h1
Nozzle:
ηN =actual KE at nozzle exit
Isentropic KE at nozzle exit=V 2
2
V 22s
=h1 − h2
h1 − h2s
7
Entropy Balance for a Closed System (Control Mass)
MER TER
CM
dWdQ
TTER
We can first perform a 1st law energy balance on the sys-tem shown above.
dU = δQ+ δW (1)
For a simple compressible system
δW = −PdV (2)
From Gibb’s equation we know
TTER dS = dU + PdV (3)
Combining (1), (2) and (3) we get
TTER dS = δQ
net in-flow dS =δQ
TTER
net out-flow dS = −δQ
TTER
From the Clausius inequality we can show that the irreversibilities in the system lead to entropyproduction and the entropy balance equation becomes
(dS)CM︸ ︷︷ ︸≡ storage
=δQ
TTER︸ ︷︷ ︸≡ entropy flow
+ dSgen︸ ︷︷ ︸≡ production
(S2 − S1)CM =Q1−2
TTER+ Sgen︸ ︷︷ ︸≥0
accumulation = (OUT − IN) + generation
Q1−2
TTER- the entropy associated with heat transfer across a
finite temperature difference, i.e. T > 0
8
Example 4-1: A bicycle tire has a volume of 1200 cm3 which is considered to be con-stant during “inflation”. Initially the tire contains air at atmospheric conditions given asP0 = 100 kPa and T0 = 20 ◦C. A student then hooks up a bicycle pump and beginsto force air from the atmosphere into the tire. After pumping stops and a new equilibrium isreached, the tire pressure is 600 kPa and the air temperature in the tire is 20 ◦C.
a) Determine the mass [kg] of air added to the tire.b) Determine the minimum amount of work [kJ ] required to reach this
end state.c) If more than the minimum work is actually used, where does this
extra energy go? That is what becomes of the extra energy?
9
Entropy Rate Balance for Open Systems (Control Volume)
TER
TER
MER
AA
B
B
FR
FR
CV
S= s m
S= -s m
B
A
1-2
1-2
1-2
1-2
1-2
1-2
1-2
1-2
1-2
B
A
A
B
A
B
B
A
A
B
S= - Qd
S= Qd
T
T
TER
TER
dQ
dQm
m
SCVdW
S=0
isolated S 0gen ≥
For the isolated system going through a process from 1→ 2
δSgen = (∆S)sys + (∆S)sur
δSgen = ∆SCV︸ ︷︷ ︸system
+
(−sAmA
1−2 + sBmB1−2 −
δQA1−2
TATER+δQB
1−2
TBTER
)︸ ︷︷ ︸
surroundings
or as a rate equation
Sgen =
(dS
dt
)CV
+
sm+Q
TTER
OUT
−
sm+Q
TTER
IN
This can be thought of as
generation = accumulation+ OUT − IN
10
Example 4-2: An adiabatic flash evaporator is used to make a small amount of clean wa-ter from dirty water. The dirty water enters as saturated liquid at 30 ◦C and at a rate of20 kg/min. Clean water leaves as a saturated vapour at 25 ◦C. Dirty water leaves theevaporator as saturated liquid at 25 ◦C.
a) Determine the mass flow rate [kg/min] of the clean water vapour.b) Determine the rate of entropy generation [kW/K] within the
evaporator.
11
Entropy Generation in a System of Components: Heat Engine
A heat engine is a device in which a working substance (control mass) undergoes a cyclic processwhile operating between two temperature reservoirs (TER).
combustion
gases
cooling
water
boiler
condenser
WP
WT
QH
QH
QL
QL
TER
TER
TH
TL
Wnet
heat
engines
isolated systems
S 0gen ≥
For the general case, the 1st law en-ergy balance gives
∆E↗0= QH −QL −Wnet = 0 (1)
The 2nd law gives
Sgen ≥ 0 for an isolated system
where Sgen = 0 implies a reversiblesystem and Sgen > 0 implies a realsystem.
An entropy balance gives
Sgen = (∆S)CM︸ ︷︷ ︸≡0 (cyclic)
+ (∆S)TER−H︸ ︷︷ ︸≡−QH/TH
+ (∆S)TER−L︸ ︷︷ ︸≡+QL/TL
+(∆S)MER↗0
Therefore
QL
QH
=TL
TH+Sgen TL
QH
(2)
Combining Eqs. (1) and (2)
Wnet = QH −QL = QH
(1−
QL
QH
)
= QH
(1−
TL
TH−Sgen TL
QH
)
= QH
(1−
TL
TH
)︸ ︷︷ ︸Wmax possible
− TLSgen︸ ︷︷ ︸Wlost due to irreversibilties
The engine efficiency is defined as the benefit over the cost
η =benefit
cost=Wnet
QH
= 1−TL
TH−TLSgen
QH
12
Conduction Heat Transfer
Reading Problems10-1→ 10-6 10-20, 10-35, 10-49, 10-54, 10-59, 10-69,
10-71, 10-92, 10-126, 10-143, 10-157, 10-16211-1→ 11-2 11-14, 11-17, 11-36, 11-41, 11-46, 11-97, 11-104
General Heat ConductionFrom a 1st law energy balance:
∂E
∂t= Qx − Qx+∆x
If the volume to the element is given asV = A ·∆x, then the mass of the element is
m = ρ ·A ·∆xx=0
xx
x+ xD
x=Linsulated
Qx
Qx+ xD
A
The energy term (KE = PE = 0) is
E = m · u = (ρ ·A ·∆x) · u
For an incompressible substance the internal energy is du = C dT and we can write
∂E
∂t= ρCA∆x
∂T
∂t
Heat flow along the x−direction is a product of the temperature difference.
Qx =kA
∆x(Tx − Tx+∆x)
where k is the thermal conductivity of the material. In the limit as ∆x→ 0
Qx = −kA∂T
∂x
This is Fourier’s law of heat conduction. The −ve in front of k guarantees that we adhere to the2nd law and that heat always flows in the direction of lower temperature.
1
We can write the heat flow rate across the differential length, ∆x as a truncated Taylor seriesexpansion as follows
Qx+∆x = Qx +∂Qx
∂x∆x
when combined with Fourier’s equation gives
Qx+∆x = −kA∂T
∂x︸ ︷︷ ︸Qx
−∂
∂x
(kA
∂T
∂x
)∆x
Noting that
Qx − Qx+∆x =∂E
∂t= ρCA∆x
∂T
∂t
By removing the common factor of A∆x we can then write the general 1-D conduction equationas
∂
∂x
(k∂T
∂x
)︸ ︷︷ ︸
longitudinalconduction
= ρC∂T
∂t︸ ︷︷ ︸thermalinertia
↙ ↘
Steady Conduction Transient Conduction
•∂T
∂t→ 0
• properties are constant
• temperature varies in a linear manner
• heat flow rate defined by Fourier’sequation
• resistance to heat flow: R =∆T
Q
• properties are constant
• therefore∂2T
∂x2=ρC
k
∂T
∂t=
1
α
∂T
∂t
where thermal diffusivity is defined as
α =k
ρC
• exact solution is complicated
• partial differential equation can besolved using approximate or graphicalmethods
2
Steady Heat Conduction
Thermal Resistance Networks
Thermal circuits based on heat flow rate, Q, temperature difference, ∆T and thermal resistance,R, enable analysis of complex systems.
Thermal Resistance
The thermal resistance to heat flow (◦C/W ) can be constructed for all heat transfer mechanisms,including conduction, convection, and radiation as well as contact resistance and spreading resis-tance.
R - film resistancef
R - fin resistancefin
R - base resistanceb
R - spreading resistancesR - contact resistancec
Tsource
Tsink
Conduction: Rcond =L
kA
Convection: Rconv =1
hA
Radiation: Rrad =1
hradA−→ hrad = εσ(T 2
s + T 2surr)(Ts + Tsurr)
Contact: Rc =1
hcA−→ hc see Table 10-2
Cartesian Systems
Resistances in Series
The heat transfer across the fluid/solid interface is based on Newton’s law of cooling
Q = hA(Tin − Tout) =Tin − ToutRconv
where Rconv =1
hA
3
The heat flow through a solid material of conductivity, k is
Q =kA
L(Tin − Tout) =
Tin − ToutRcond
where Rcond =L
kA
By summing the temperature drop acrosseach section, we can write:
Q R1 = (T∞1 − T1)
Q R2 = (T1 − T2)
Q R3 = (T2 − T3)
Q R4 = (T3 − T∞2)
Q
(4∑i=1
Ri
)= (T∞1 − T∞2)
The total heat flow across the system can be written as
Q =T∞1 − T∞2
Rtotal
where Rtotal =4∑i=1
Ri
4
Resistances in Parallel
For systems of parallel flow paths asshown above, we can use the 1st lawto preserve the total energy
Q = Q1 + Q2
where we can write
L
T1
T2
R1
R2
R3
k1
k2
k3
Q1 =T1 − T2
R1
R1 =L
k1A1
Q2 =T1 − T2
R2
R2 =L
k2A2
Q =∑Qi = (T1 − T2)
(∑ 1
Ri
)where
1
Rtotal
=∑ 1
Ri
= UA
In general, for parallel networks we can use a parallel resistor network as follows:
T1
T1
T2 T
2
R1
RtotalR
2
R3
=
1
Rtotal
=1
R1
+1
R2
+1
R3
+ · · ·
and
Q =T1 − T2
Rtotal
5
Thermal Contact Resistance
• real surfaces have microscopic roughness,leading to non-perfect contacts where
– 1 - 4% of the surface area is in solid-solidcontact, the remainder consists of air gaps
• the total heat flow rate can bewritten as
Qtotal = hcA∆Tinterface
where:
hc = thermal contact conductanceA = apparent or projected area of the contact∆Tinterface = average temperature drop across the interface
The conductance, hc and the contact resistance,Rc can be written as
hcA =Qtotal
∆Tinterface=
1
Rc
Table 10-2 can be used to obtain some representative values for contact conductance
Table 10-2: Contact Conductances
6
Cylindrical Systems
L
r1
r2
Qr
T1
T2
A=2 rLp
k
r
Steady, 1D heat flow from T1 to T2
in a cylindrical system occurs in a ra-dial direction where the lines of con-stant temperature (isotherms) are con-centric circles, as shown by the dottedline and T = T (r).Performing a 1st law energy balanceon a control mass from the annularring of the cylindrical cylinder gives:
Qr =T1 − T2(
ln(r2/r1)
2πkL
) where R =
(ln(r2/r1)
2πkL
)
Composite Cylinders
Then the total resistance can be written as
Rtotal = R1 +R2 +R3 +R4
=1
h1A1
+ln(r2/r1)
2πk2L+
ln(r3/r2)
2πk3L+
1
h4A4
7
Example 5-1: Determine the temperature (T1) of an electric wire surrounded by a layer ofplastic insulation with a thermal conductivity if 0.15 W/mK when the thickness of the insu-lation is a) 2 mm and b) 4 mm, subject to the following conditions:
Given: Find:I = 10 A T1 = ???
∆ε = ε1 − ε2 = 8 V when:D = 3 mm δ = 2 mmL = 5 m δ = 4 mmk = 0.15 W/mK
T∞ = 30 ◦Ch = 12 W/m2 ·K
Critical Radius of Insulation
Consider a steady, 1-D problem where an insulationcladding is added to the outside of a tube with constantsurface temperature Ti. What happens to the heat trans-fer as insulation is added, i.e. we increase the thicknessof the insulation?
The resistor network can be written as a series combina-tion of the resistance of the insulation, R1 and the con-vective resistance,R2
Rtotal = R1 +R2 =ln(ro/ri)
2πkL+
1
h2πroL
Could there be a situation in which adding insulation increases the overall heat transfer?
8
dRtotal
dro=
1
2πkroL−
1
h2πr2oL
= 0 ⇒ rcr,cyl =k
h[m]
resis
tan
ce
Rtotal
R1
added insulation
lowers resistance
added insulation
increases resistance
Rbare
R2
ri
ror
c
There is always a value of rcr,cal, but there is a minimum in heat transfer only if rcr,cal > ri
Spherical Systems
For steady, 1D heat flow in spherical geometries we can writethe heat transfer in the radial direction as
Q =4πkriro
(r0 − ri)(Ti − To) =
(Ti − To)R
where: R =ro − ri4πkriro
ri
ro
Ti
To
The critical radius of insulation for a spherical shell is given as
rcr,sphere =2k
h[m]
9
Heat Transfer from Finned SurfacesWe can establish a 1st law balance over the thin slice ofthe fin between x and x+ ∆x such that
Qx − Qx+∆x − P∆x︸ ︷︷ ︸Asurface
h(T − T∞) = 0
From Fourier’s law we know
Qx − Qx+∆x = kAc
d2T
dx2∆x
Therefore the conduction equation for a fin with constantcross section is
kAc
d2T
∂x2︸ ︷︷ ︸longitudinalconduction
−hP (T − T∞)︸ ︷︷ ︸lateral
convection
= 0
Let the temperature difference between the fin and the surroundings (temperature excess) beθ = T (x)− T∞ which allows the 1-D fin equation to be written as
d2θ
dx2−m2θ = 0 where m =
(hP
kAc
)1/2
The solution to the differential equation for θ is
θ(x) = C1 sinh(mx) + C2 cosh(mx) [≡ θ(x) = C1emx + C2e
−mx]
Potential boundary conditions include:
Base: → @x = 0 θ = θbTip: → @x = L θ = θL [T -specified tip]
θ =dθ
dx
∣∣∣∣∣x=L
= 0 [adiabatic (insulated) tip]
θ → 0 [infinitely long fin]
Substituting the boundary conditions to find the constants of integration, the temperature distribu-tion and fin heat transfer rate can be determined as follows:
Case 1: Prescribed temperature (θ@ x+L = θL)
θ(x)
θb=
(θL/θb) sinhmx+ sinhm(L− x)
sinhmL
10
Qb = M(coshmL− θL/θb)
sinhmL
Case 2: Adiabatic tip(dθ
dx
∣∣∣∣∣x=L
= 0
)
θ(x)
θb=
coshm(L− x)
coshmLQb = M tanhmL
Case 3: Infinitely long fin (θ → 0)
θ(x)
θb= e−mx Qb = M
where
m =√hP/(kAc)
M =√hPkAc θb
θb = Tb − T∞
Fin Efficiency and Effectiveness
The dimensionless parameter that compares the actual heat transfer from the fin to the ideal heattransfer from the fin is the fin efficiency
η =actual heat transfer rate
maximum heat transfer rate whenthe entire fin is at Tb
=Qb
hPLθb
If the fin has a constant cross section then
η =tanh(mL)
mL
An alternative figure of merit is the fin effectiveness given as
εfin =total fin heat transfer
the heat transfer that would haveoccurred through the base area
in the absence of the fin
=Qb
hAcθb
11
How to Determine the Appropriate Fin Length
• theoretically an infinitely long fin willdissipate the most heat
• but practically, an extra long fin is in-efficient given the exponential temper-ature decay over the length of the fin
• so what is a realistic fin length in orderto optimize performance and cost
If we determine the ratio of heat flow for afin with an insulated tip (Case 2) versus aninfinitely long fin (Case 3) we can assess therelative performance of a conventional fin
QCase 2
QCase 3
=M tanhmL
M= tanhmL
High
heat
transfer
Low
heat
transfer
No
heat
transfer
T
T
T
x
Tb
L
T = highT= low
T = 0
h, T
T(x)
0
Transient Heat ConductionPerforming a 1st law energy balance on a planewall gives
Qcond =TH − TsL/(k ·A)
= Qconv =Ts − T∞1/(h ·A)
where the Biot number can be obtained as follows:
TH − TsTs − T∞
=L/(k ·A)
1/(h ·A)=
internal resistance to H.T.external resistance to H.T.
=hL
k= Bi
Rint << Rext: the Biot number is small and we can conclude
TH − Ts << Ts − T∞ and in the limit TH ≈ Ts
Rext << Rint: the Biot number is large and we can conclude
Ts − T∞ << TH − Ts and in the limit Ts ≈ T∞
12
Transient Conduction Analysis• if the internal temperature of a body remains relatively constant with respect to time
– can be treated as a lumped system analysis
– heat transfer is a function of time only, T = T (t)
T T T T
T(t)
L LL L
tt
x x
T(x,0) = Ti
T(x,t)
Bi 0.1≤ Bi > 0.1
Bi ≤ 0.1: temperature profile is not a function of positiontemperature profile only changes with respect to time→ T = T (t)use lumped system analysis
Bi > 0.1: temperature profile changes with respect to time and position→ T = T (x, t)use approximate analytical or graphical solutions (Heisler charts)
Lumped System Analysis
At t > 0, T = T (x, y, z, t), however, whenBi ≤ 0.1 then we can assume T ≈ T (t).
13
Performing a 1st law energy balance on the control volume shown below
dEC.M.
dt= Ein − Eout + Eg↗0
If we assume PE andKE to be negligible then
dU
dt= −Q ⇐
dU
dt< 0 implies U is decreasing
For an incompressible substance specific heat is constant and we can write
mC︸ ︷︷ ︸≡Cth
dT
dt= − Ah︸︷︷︸
1/Rth
(T − T∞)
where Cth = lumped capacitance
CthdT
dt= −
1
Rth
(T − T∞)
We can integrate and apply the initial condition, T = Ti @t = 0 to obtain
T (t)− T∞Ti − T∞
= e−t/(Rth·Cth) = e−t/τ = e−bt
where
1
b= τ
= Rth · Cth
= thermal time constant
=mC
Ah=ρV C
Ah
The total heat transferred over the time period 0→ t∗ is
Qtotal = mC(Ti − T∞)[1− e−t∗/τ ]
14
Example 5-2: Determine the time it takes a fuse to melt if a current of 3 A suddenly flowsthrough the fuse subject to the following conditions:
Given:D = 0.1 mm Tmelt = 900 ◦C k = 20 W/mK
L = 10 mm T∞ = 30 ◦C α = 5× 10−5 m2/s ≡ k/ρCp
Assume:
• constant resistance R = 0.2 ohms
• the overall heat transfer coefficient is h = hconv + hrad = 10 W/m2K
• neglect any conduction losses to the fuse support
15
Approximate Analytical and Graphical Solutions (Heisler Charts)IfBi > 0.1
• need to solve the partial differential equation for temperature
• leads to an infinite series solution⇒ difficult to obtain a solution(see pp. 481 - 483 for exact solution by separation of variables)
We must find a solution to the PDE
∂2T
∂x2=
1
α
∂T
∂t⇒
T (x, t)− T∞Ti − T∞
=∞∑
n=1,3,5...
Ane
(−λn
L
)2
αt
cos
(λnx
L
)
By using dimensionless groups, we can reduce the temperature dependence to 3 dimensionlessparameters
Dimensionless Group Formulation
temperature θ(x, t) =T (x, t)− T∞Ti − T∞
position X = x/L
heat transfer Bi = hL/k Biot number
time Fo = αt/L2 Fourier number
note: Cengel uses τ instead of Fo.
Now we can write
θ(x, t) = f(X,Bi, Fo)
The characteristic length for the Biot number is
slab L = Lcylinder L = rosphere L = ro
contrast this versus the characteristic length for the lumped system analysis.
16
With this, two approaches are possible
1. use the first term of the infinite series solution. This method is only valid for Fo > 0.2
2. use the Heisler charts for each geometry as shown in Figs. 11-15, 11-16 and 11-17
First term solution: Fo > 0.2→ error about 2% max.
Plane Wall: θwall(x, t) =T (x, t)− T∞Ti − T∞
= A1e−λ2
1Fo cos(λ1x/L)
Cylinder: θcyl(r, t) =T (r, t)− T∞Ti − T∞
= A1e−λ2
1Fo J0(λ1r/ro)
Sphere: θsph(r, t) =T (r, t)− T∞Ti − T∞
= A1e−λ2
1Fosin(λ1r/ro)
λ1r/ro
λ1, A1 can be determined from Table 11-2 based on the calculated value of the Biot number (willlikely require some interpolation). The Bessel function, J0 can be calculated using Table 11-3.
Using Heisler Charts
• find T0 at the center for a given time (Table 11-15 a, Table 11-16 a or Table 11-17 a)
• find T at other locations at the same time (Table 11-15 b, Table 11-16 b or Table 11-17 b)
• findQtot up to time t (Table 11-15 c, Table 11-16 c or Table 11-17 c)
Example 5-3: An aluminum plate made of Al 2024-T6 with a thickness of 0.15 m isinitially at a temperature of 300 K. It is then placed in a furnace at 800 K with aconvection coefficient of 500 W/m2K.
Find: i) the time (s) for the plate midplane to reach 700 Kii) the surface temperature at this condition. Use both the Heisler charts
and the approximate analytical, first term solution.
17
Convection Heat Transfer
Reading Problems12-1→ 12-8 12-40, 12-49, 12-68, 12-70, 12-87, 12-9813-1→ 13-6 13-39, 13-47, 13-5914-1→ 14-4 14-18, 14-24, 14-45, 14-82
Introduction
↙ ↘Newton’s Law of Cooling
Qconv =∆T
Rconv
= hA(Tw − T∞)
⇒ Rconv =1
hA
Typical Values of h (W/m2K)
Natural gases: 3-20Convection water: 60 - 900
Forced gases: 30 - 300Convection oils: 60 - 1800
water: 100 -1500
Boiling water: 3000 - 105
Condensation steam: 3000 - 105
Controlling Factors
Geometry: shape, size, aspect ratio and orientationFlow Type: forced, natural, laminar, turbulent,
internal, externalBoundary: isothermal (Tw = constant) or
isoflux (qw = constant)Fluid Type: viscous oil, water, gases or liquid metalsProperties: all properties determined at film temperature
Tf = (Tw + T∞)/2Note: ρ and ν ∝ 1/Patm ⇒ see Q12-40density: ρ ((kg/m3)specific heat: Cp (J/kg ·K)
dynamic viscosity: µ, (N · s/m2)
kinematic viscosity: ν ≡ µ/ρ (m2/s)
thermal conductivity: k, (W/m ·K)
thermal diffusivity: α, ≡ k/(ρ · Cp) (m2/s)
Prandtl number: Pr, ≡ ν/α (−−)
volumetric compressibility: β, (1/K)
1
Forced ConvectionThe simplest forced convection configuration to consider is the flow of mass and heat near a flatplate as shown below.
• flow forms thin layers that can slip past one another at different velocities
• as Reynolds number increases the flow has a tendency to become more chaotic resulting indisordered motion known as turbulent flow
– transition from laminar to turbulent is called the critical Reynolds number,Recr
Recr =U∞xcr
ν
– for flow over a flat plateRecr ≈ 500, 000
– x < xcr the boundary layer is laminar; x > xcr the boundary layer is turbulent
Boundary Layers
Velocity Boundary Layer
• the region of fluid flow over the plate where viscous effects dominate is called the velocityor hydrodynamic boundary layer
• the velocity at the surface of the plate, y = 0, is set to zero, U@y=0 = 0 m/s because ofthe no slip condition at the wall
2
• the velocity of the fluid progressively increases away from the wall until we reach approxi-mately 0.99 U∞ which is denoted as the δ, the velocity boundary layer thickness.
• the region beyond the velocity boundary layer is denoted as the inviscid flow region, wherefrictional effects are negligible and the velocity remains relatively constant at U∞
Thermal Boundary Layer
• the thermal boundary layer is arbitrarily selected as the locus of points where
T − TwT∞ − Tw
= 0.99
• for low Prandtl number fluids the velocity boundary layer is fully contained within the ther-mal boundary layer
• conversely, for high Prandtl number fluids the thermal boundary layer is contained withinthe velocity boundary layer
Flow Over Plates
1. Laminar Boundary Layer Flow, Isothermal (UWT)
All laminar formulations forRe < 500, 000. The local value of the Nusselt number is given as
Nux = 0.332Re1/2x Pr1/3 ⇒ local, laminar, UWT, Pr ≥ 0.6
An average value of the heat transfer coefficient for the full extent of the plate can be obtained byusing the mean value theorem.
NuL =hLL
kf= 0.664 Re
1/2L Pr1/3 ⇒ average, laminar, UWT, Pr ≥ 0.6
For low Prandtl numbers, i.e. liquid metals
Nux = 0.565Re1/2x Pr1/2 ⇒ local, laminar, UWT, Pr ≤ 0.6
2. Turbulent Boundary Layer Flow, Isothermal (UWT)
All turbulent formulations for 500, 000 ≤ Re ≤ 107. The local Nusselt number is given as
Nux = 0.0296Re0.8x Pr1/3 ⇒
local, turbulent, UWT,0.6 < Pr < 60
3
and the average Nusselt number is
NuL = 0.037Re0.8L Pr1/3 ⇒
average, turbulent, UWT,0.6 < Pr < 60
3. Combined Laminar and Turbulent Boundary Layer Flow, Isothermal (UWT)
When (Tw − T∞) is constant
hL =1
L
∫ L0hdx =
1
L
{∫ xcr
0hlamx dx+
∫ Lxcr
hturx dx
}
NuL =hLL
k= (0.037Re0.8
L − 871) Pr1/3 ⇒
average, combined, UWT,0.6 < Pr < 60,500, 000 ≤ ReL ≤ 107
4. Laminar Boundary Layer Flow, Isoflux (UWF)
Nux = 0.453Re1/2x Pr1/3 ⇒ local, laminar, UWF, Pr ≥ 0.6
5. Turbulent Boundary Layer Flow, Isoflux (UWF)
Nux = 0.0308Re4/5x Pr1/3 ⇒ local, turbulent, UWF, Pr ≥ 0.6
Example 6-1: Hot engine oil with a bulk temperature of 60 ◦C flows over a horizontal, flatplate 5 m long with a wall temperature of 20 ◦C. If the fluid has a free stream velocity of2 m/s, determine the heat transfer rate from the oil to the plate if the plate is assumed to beof unit width.
4
Flow Over Cylinders and Spheres
1. Boundary Layer Flow Over Circular Cylinders, Isothermal (UWT)
The Churchill-Bernstein (1977) correlation for the average Nusselt number forlong (L/D > 100) cylinders is
NuD = 0.3 +0.62Re
1/2D Pr1/3
[1 + (0.4/Pr)2/3]1/4
1 +
(ReD
282, 000
)5/84/5
⇒ average, UWT,ReD < 107, 0 ≤ Pr ≤ ∞, ReD · Pr > 0.2
All fluid properties are evaluated at Tf = (Tw + T∞)/2.
2. Boundary Layer Flow Over Non-Circular Cylinders, Isothermal (UWT)
The empirical formulations of Zhukauskas and Jakob given in Table 12-3 are commonly used,where
NuD ≈hD
k= C RemD Pr
1/3 ⇒ see Table 12-3 for conditions
3. Boundary Layer Flow Over a Sphere, Isothermal (UWT)
For flow over an isothermal sphere of diameterD, Whitaker recommends
NuD = 2 +[0.4Re
1/2D + 0.06Re
2/3D
]Pr0.4
(µ∞
µs
)1/4
⇒
average, UWT,0.7 ≤ Pr ≤ 380
3.5 < ReD < 80, 000
where the dynamic viscosity of the fluid in the bulk flow, µ∞ is based on the free stream tem-perature, T∞ and the dynamic viscosity of the fluid at the surface, µs, is based on the surfacetemperature, Ts. All other properties are based on T∞.
5
Example 6-2: An electric wire with a 1 mm diameter and a wall temperature of 325 Kis cooled by air in cross flow with a free stream temperature of 275 K. Determine the airvelocity required to maintain a steady heat loss per unit length of 70 W/m.
Internal Flow
Lets consider fluid flow in a duct bounded by a wall that is at a different temperature than the fluid.For simplicity we will examine a round tube of diameterD as shown below
The Reynolds number is given as: ReD =UmD
ν. For flow in a tube:
ReD < 2300 laminar flow
2300 < ReD < 4000 transition to turbulent flow
ReD > 4000 turbulent flow
6
For engineering calculations, we typically assume thatRecr ≈ 2300, therefore
ReD
{< Recr laminar> Recr turbulent
Hydrodynamic (Velocity) Boundary Layer
• when the boundary layer grows to the tube radius, r, the boundary layers merge
– this flow length is called the flow entrance length, Lh– 0 ≤ x ≤ Lh is the hydrodynamic entrance region
– Lh < x ≤ L is the fully developed region or hydrodynamically developed region
• the hydrodynamic boundary layer thickness can be approximated as
δ(x) ≈ 5x
(Umx
ν
)−1/2
=5x√Rex
• the hydrodynamic entry length can be approximated as
Lh ≈ 0.05ReDD (laminar flow)
Thermal Boundary Layer
• a thermal entrance region develops from 0 ≤ x ≤ Lt
• the thermal entry length can be approximated as
Lt ≈ 0.05ReDPrD = PrLh (laminar flow)
• for turbulent flow Lh ≈ Lt ≈ 10D
7
Wall Boundary Conditions
1. Uniform Wall Heat Flux: The total heat transfer from the wall to the fluid stream can be deter-mined by performing an energy balance over the tube. If we assume steady flow conditions,min = mout = m then the energy balance becomes
Q = qwA = m(hout − hin) = mCp(Tout − Tin)
Since the wall flux qw is uniform, the localmean temperature is linear with x.
Tm,x = Tm,i +qwA
mCp
The surface temperature can be determinedfrom
Tw = Tm +qw
h
2. Isothermal Wall: Using Newton’s law of cooling we can determine the average rate of heattransfer to or from a fluid flowing in a tube
Q = hA (Tw − Tm)︸ ︷︷ ︸average ∆T
From an energy balance over a control volume in the fluid, we can determine
Q = mCpdTm
Equating the two equations above we find
mCpdTm = hA (Tw − Tm)︸ ︷︷ ︸average ∆T
By isolating the temperature terms and inte-grating we obtain
ln
(Tw − ToutTw − Tin
)= −
hA
mCp
Because of the exponential temperature decay within the tube, it is common to present themean temperature from inlet to outlet as a log mean temperature difference where
∆Tln =Tout − Tin
ln
(Tw − ToutTw − Tin
) =Tout − Tin
ln(∆Tout/∆Tin)⇒ Q = hA∆Tln
8
1. Laminar Flow in Circular Tubes, Isothermal (UWT) and Isoflux (UWF)
For laminar flow whereReD ≤ 2300
NuD = 3.66 ⇒ fully developed, laminar, UWT, L > Lt & Lh
NuD = 4.36 ⇒ fully developed, laminar, UWF, L > Lt & Lh
NuD = 1.86
(ReDPrD
L
)1/3 (µbµs
)0.14
⇒
developing laminar flow, UWT,Pr > 0.5
L < Lh or L < Lt
For non-circular tubes the hydraulic diameter, Dh = 4Ac/P can be used in conjunction withTable 13-1 to determine the Reynolds number and in turn the Nusselt number.
2. Turbulent Flow in Circular Tubes, Isothermal (UWT) and Isoflux (UWF)
For turbulent flow whereReD ≥ 2300 the Dittus-Boelter equation (Eq. 13-68) can be used
NuD = 0.023Re0.8D Prn ⇒
turbulent flow, UWT or UWF,0.7 ≤ Pr ≤ 160
ReD > 2, 300
n = 0.4 heatingn = 0.3 cooling
For non-circular tubes, again we can use the hydraulic diameter,Dh = 4Ac/P to determine boththe Reynolds and the Nusselt numbers.
In all cases the fluid properties are evaluated at the mean fluid temperature given as
Tmean =1
2(Tm,in + Tm,out)
except for µs which is evaluated at the wall temperature, Ts.
9
Natural Convection
What Drives Natural Convection?• fluids tend to expand when heated and contract
when cooled at constant pressure
• therefore a fluid layer adjacent to a surface willbecome lighter if heated and heavier if cooledby the surface
• a lighter fluid will flow upward and a coolerfluid will flow downward
• as the fluid sweeps the wall, heat transfer willoccur in a similar manner to boundary layerflow however in this case the bulk fluid is sta-tionary as opposed to moving at a constant ve-locity in the case of forced convection
In natural convection, the Grashof number is analogous to the Reynolds number.
Gr =buouancy forceviscous force
=gβ(Tw − T∞)L3
ν2
Natural Convection Over Surfaces
• natural convection heat transferdepends on geometry and ori-entation
• note that unlike forced convec-tion, the velocity at the edge ofthe boundary layer goes to zero
• the velocity and temperatureprofiles within a boundary layerformed on a vertical plate ina stationary fluid looks as fol-lows:
10
Natural Convection Heat Transfer Correlations
The general form of the Nusselt number for natural convection is as follows:
Nu = f(Gr, Pr) ≡ CGrmPrn where Ra = Gr · Pr
• C depends on geometry, orientation, type of flow, boundary conditions and choice of char-acteristic length.
• m depends on type of flow (laminar or turbulent)
• n depends on the type of fluid and type of flow
• Table 14-1 should be used to find Nusselt number for various combinations of geometry andboundary conditions
– for ideal gases β = 1/Tf , (1/K)
– all fluid properties are evaluated at the film temperature, Tf = (Tw + T∞)/2.
1. Laminar Flow Over a Vertical Plate, Isothermal (UWT)
The general form of the Nusselt number is given as
NuL =hLkf
= C
gβ(Tw − T∞)L3
ν2︸ ︷︷ ︸≡Gr
1/4 ν
α︸︷︷︸≡Pr
1/4
= C Gr1/4L Pr1/4︸ ︷︷ ︸Ra1/4
where
RaL = GrLPr =gβ(Tw − T∞)L3
αν
2. Laminar Flow Over a Long Horizontal Circular Cylinder, Isothermal (UWT)
The general boundary layer correlation is
NuD =hD
kf= C
gβ(Tw − T∞)D3
ν2︸ ︷︷ ︸≡Gr
1/4 ν
α︸︷︷︸≡Pr
1/4
= C Gr1/4D Pr1/4︸ ︷︷ ︸Ra
1/4D
where
RaD = GrDPr =gβ(Tw − T∞)L3
αν
11
Natural Convection From Plate Fin Heat Sinks
The average Nusselt number for an isothermal plate fin heat sink with natural convection can bedetermined using
NuS =hS
kf=
[576
(RaSS/L)2+
2.873
(RaSS/L)0.5
]−0.5
Two factors must be considered in the selection of thenumber of fins
• more fins results in added surface area and re-duced boundary layer resistance,
R ↓=1
hA ↑
• more fins leads to a decrease fin spacing and adecrease in the heat transfer coefficient
R ↑=1
h ↓ A
A basic optimization of the fin spacing can be obtained as follows:
Q = hA(Tw − T∞)
where the fins are assumed to be isothermal and the surface area is 2nHL, with the area of the finedges ignored.
For isothermal fins with t < S
Sopt = 2.714
(L
Ra1/4L
)
with
RaL =gβ(Tw − T∞)L3
ν2Pr
The corresponding value of the heat transfer coefficient is
h = 1.307kf/Sopt
All fluid properties are evaluated at the film temperature.
12
Example 6-3: Find the optimum fin spacing, Sopt and the rate of heat transfer, Q for thefollowing plate fin heat sink cooled by natural convection.
Given:
W = 120 mm H = 24 mmL = 18 mm t = 1 mmTw = 80 ◦C T∞ = 25 ◦CP∞ = 1 atm fluid = air
Find: Sopt and the corresponding heat transfer, Q
13
Radiation Heat Transfer
Reading Problems15-1→ 15-7 15-27, 15-33, 15-50, 15-57, 15-77, 15-79,
15-96, 15-107, 15-108
IntroductionA narrower band inside the thermal radiation spectrum is denoted as the visible spectrum, that isthe thermal radiation that can be seen by the human eye. The visible spectrum occupies roughly0.4− 0.7 µm. Thermal radiation is mostly in the infrared range. As objects heat up, their energylevel increases, their frequency, ν, increases and the wavelength of the emitted radiation decreases.
10 10 10 10 10 10 10 10 10 10-5 -4 -3 -2 -1 0 1 2 3 4
m
Visible
Gamma rays
X rays
Ultraviolet
Infrared
Thermal radiation
Microwave
,
vio
let
blu
e
gre
en
yello
w
red
Blackbody RadiationA blackbody is an ideal radiator that absorbs all incident radiation regardless of wavelength anddirection.
Definitions1. Blackbody emissive power: the radiation emitted by a blackbody per unit time and per
unit surface area
Eb = σ T 4 [W/m2] ⇐ Stefan-Boltzmann law
where Stefan-Boltzmann constant = 5.67× 10−8 W/(m2 ·K4) and the temperature Tis given inK.
1
2. Spectral blackbody emissive power: the amount of radiation energy emitted by a black-body per unit surface area and per unit wavelength about the wavelength λ. The followingrelationship between emissive power, temperature and wavelength is known asPlank’s distribution law
Eb,λ =C1
λ5[exp(C2/λT )− 1][W/(m2 · µm)]
where
C1 = 2πhC20 = 3.74177× 108 [W · µm4/m2]
C2 = hC0/K = 1.43878× 104 [µm ·K]
The wavelength at which the peak emissive power occurs for a given temperature can beobtained from Wien’s displacement law
(λT )max power = 2897.8 µm ·K
3. Blackbody radiation function: the fraction of radiation emitted from a blackbody at tem-perature, T in the wavelength band λ = 0→ λ
f0→λ =
∫ λ0Eb,λ(T ) dλ∫ ∞
0Eb,λ(T ) dλ
=
∫ λ0
C1
λ5[exp(C2/λT )− 1]dλ
σT 4
let t = λT and dt = T dλ, then
f0→λ =
∫ t0
C1T5(1/T )dt
t5[exp(C2/t)− 1]
σT 4
=C1
σ
∫ λT0
dt
t5[exp(C2/t)− 1]
= f(λT )
f0→λ is tabulated as a function λT in Table 15.2
We can easily find the fraction of radiation emitted by a blackbody at temperature T over adiscrete wavelength band as
fλ1→λ2 = f(λ2T )− f(λ1T )
fλ→∞ = 1− f0→λ
2
Radiation Properties of Real Surfaces
The thermal radiation emitted by a real surface is afunction of surface temperature, T , wavelength, λ,direction and surface properties.
Eλ = f(T, λ, direction, surface properties)
⇒ spectral emissive power
while for a blackbody, the radiation was only a function of temperature and wavelength
Eb,λ = f(T, λ) → diffuse emitter ⇒ independent of direction
Definitions
1. Diffuse surface: properties are independent of direction.
2. Gray surface: properties are independent of wavelength.
3. Emissivity: defined as the ratio of radiation emitted by a surface to the radiation emitted bya blackbody at the same surface temperature.
ε(T ) =radiation emitted by surface at temperature T
radiation emitted by a black surface at T
=
∫ ∞0Eλ(T ) dλ∫ ∞
0Ebλ(T ) dλ
=
∫ ∞0ελ(T )Ebλ(T ) dλ
Eb(T )=E(T )
σT 4
where ε changes rather quickly with surface temperature.
Typical Emissivity Valuesmetal (polished) ε ≈ 0.1metal (oxidized) ε ≈ 0.3− 0.4skin ε ≈ 0.9graphite ε ≈ 0.95
3
4. Irradiation,G: the radiation energy incident on a surface per unit area and per unit time
An energy balance based on incident radiation gives
G = ρG+ αG+ τG
where
ρ = reflectivityα = absorptivityτ = transmissivity
⇒ function of λ& T of the incident radiation G
ε = emissivity ⇒ function of λ& T of the emitting surface
If we normalize with respect to the total irradiation
α+ ρ+ τ = 1
In general ε 6= α. However, for a diffuse-gray surface (properties are independent of wave-length and direction)
ε = α diffuse-gray surface
5. Radiosity, J : the total radiation energy leaving a surface per unit area and per unit time.
For a surface that is gray and opaque, i.e. ε = α and α+ ρ = 1, the radiosity is given as
J = radiation emitted by the surface + radiation reflected by the surface
= ε Eb + ρG
= εσT 4 + ρG
Since ρ = 0 for a blackbody, the radiosity of a blackbody is
J = σT 4
4
Diffuse-Gray Surfaces, ε = α
Kirchhoff’s Law
The absorptivity, α(λ, T, direction) of a non-black surface is always equal to the emissivity,ε(λ, T, direction) of the same surface when the surface is in thermal equilibrium with the radi-ation that impinges on it.
ε(λ, T, φ, θ) = α(λ, T, φ, θ)
To a lesser degree of certainty we can write a more restrictive form of Kirchhoff’s law for diffuse-gray surfaces where
ε(T ) = α(T )
While Kirchhoff’s law requires that the radiant source and the surface be in thermal equilibrium,this is seldom the case. The law can still be used but you should proceed with caution when thetwo temperatures differ by more than 100K.
View Factor (Shape Factor, Configuration Factor)
• Definition: The view factor, Fi→jis defined as the fraction of radiationleaving surface i which is interceptedby surface j. Hence
Fi→j =Qi→j
AiJi=
radiation reaching jradiation leaving i
Fi→j =1
Ai
∫Ai
∫Aj
cos θi cos θj
πR2dAjdAi
Fj→i =1
Aj
∫Aj
∫Ai
cos θi cos θj
πR2dAidAj
5
View Factor Relations
Reciprocity Relation
The last two equations show that
AiFi→j = AjFj→i
Summation Relation
A1J1 = Q1→1 + Q1→2 + . . .+ Q1→N
Therefore
1 =N∑j=1
Qi→j
AiJi
=N∑j=1
Fi→j
Hence
N∑j=1
Fi→j = 1 ; i = 1, 2, . . . , N
Note that Fi→i 6= 0 for a concave surface. For a plane or convex surface Fi→i = 0.
Superposition Relation
If the surface is not available in the tables sometimesit can be treated as the sum of smaller known surfacesto form the full extent of the surface of interest.
F1→(2,3) = F1→2 + F1→3
Symmetry Relation
If the problem is symmetric, then the view factors will also be symmetric.
6
Hottel Crossed String Method
Can be applied to 2D problems where surfaces are any shape, flat, concave or convex. Note for a2D surface the area,A is given as a length times a unit width.
⇒
A1F12 = A2F21 =(total crossed)− (total uncrossed)
2
A1 andA2 do not have to be parallel
A1F12 = A2F21 =1
2[(ac+ bd)︸ ︷︷ ︸
crossed
− (bc+ ad)︸ ︷︷ ︸uncrossed
]
Radiation Exchange Between SurfacesIn general, radiation exchange between surfaces should include:
• irradiation of each surface accounting for all energy reflected from other surfaces
• multiple reflections may occur before all energy is absorbed
Diffuse-Gray Surfaces Forming an Enclosure
To help simplify radiation analyses in diffuse, gray enclosures we will assume
1. each surface of the enclosure is isothermal
7
2. radiosity, Ji, and irradiation,Gi are uniform over each surface
3. the surfaces are opaque (τi = 0) and diffuse-gray (αi = εi)
4. the cavity is filled with a fluid which does not participate in the radiative exchange process
Radiation Heat Transfer To or From a Surface
• an energy balance on the i′th surface gives:
Qi = qiAi = Ai(Ji −Gi)
recasting the energy balance:
Qi = Ai[(Ei + ρiGi)︸ ︷︷ ︸Ji
− (ρiGi + αiGi)︸ ︷︷ ︸Gi
] = Ai(Ei − αiGi) (1)
8
where:
Ji = Ei + ρiGi (2) ⇒ radiosity
Ei = εiEb,i = εiσT4i (3) ⇒ emmisive power
ρi = 1− αi = 1− εi (4) ⇒ since αi + ρi + τi↗0= 1
and αi = εi
Combining Eqs. 2, 3 and 4 gives
Ji = εiEb,i + (1− εi)Gi (5)
Combining this with Eq. 1 gives the net radiation heat transfer to or from surface “i”
Qi =Eb,i − Ji(
1− εiεiAi
) ≡ potential differencesurface resistance
this surface resistance represents real surface behavior
Note: for a black surface
εi = αi = 1
and Eq. 5 becomes
Ji = Eb,i = σT 4i ⇐
Radiation Heat Transfer Between Surfaces• by inspection it is clearly seen that
{irradiation on
surface i
}=
{radiation leaving theremaining surfaces
}
AiGi =N∑j=1
Fj→i(AjJj) =N∑j=1
AiFi→jJi ⇐ (from reciprocity)
9
Therefore
Gi =N∑j=1
Fi→jJj
Combining this with Eq. 5 gives
Ji = εi σT4i︸ ︷︷ ︸
Eb,i
+(1− εi)N∑j=1
Fi→jJj
In addition by performing an energy balance at surface “i”, we can write
Qi = energy out− energy in
= AiJi −N∑j=1
AiFi→jJj
Since the summation rule statesN∑j=1
Fi→j = 1, the above equation becomes
Qi = Ai
N∑j=1
Fi→j︸ ︷︷ ︸≡1
Ji −N∑j=1
Fi→jJj
Qi =
N∑j=1
AiFi→j(Ji − Jj)
or
Qi =N∑j=1
Ji − Jj(1
AiFi→j
) ≡ potential differencespace resistance
• the space resistance can be used for any gray, diffuse and opaque surfaces that form anenclosure
10
Radiation Exchange in EnclosuresThe Two-Surface Enclosure
• radiation from surface 1 must equalradiation to surface 2
Q1 = −Q2 = Q12
• the resistor network will consist of2 surface resistances and 1 spaceresistance
• the net radiation exchange can bedetermined as follows:
Q1
Q2
A , T ,1 1 1
A , T ,2 2 2
Q12
Q12 =Eb,1 − Eb,2Rtotal
=σ(T 4
1 − T 42 )
1− ε1
ε1A1
+1
A1F12
+1− ε2
ε2A2
11
The Three-Surface Enclosure
• radiative heat transfer betweenall combinations of surfacesmust be accounted for
• the resistor network will consistof 3 surface resistances and 3space resistances
• leads to a system of 3 equationsin 3 unknowns
• the algebraic sum of the cur-rents (net radiation transfer) ateach node must equal zero.Note: this assumes all heat flowis into the node.
• if the assumed direction of cur-rent flow is incorrect, your willget a -ve value of Q
Performing an energy balance at eachnode:
Q
RR
R
R
R R
Q
Q
Q
Q
Q
Q
Q
Q
E EJ J
J
b1 b21
21
3
2
3
12
1323
2
12
12
13
13
23
23
3
1
1-
1-
1
1 1
1-
A1
, A , T1 1
, A , T2 2
, A , T3 3
A3
A2
1
3
2
1
1
2
3
3
2A F1 12
A F1 13 A F2 23 nodes for applying
surface energy balance
at J1 ⇒ Q1 + Q12 + Q13 = 0
Eb1 − J1
1− ε1
ε1A1
+J2 − J1
1
A1F12
+J3 − J1
1
A1F13
= 0 (5)
at J2 ⇒ Q12 + Q2 + Q23 = 0
J1 − J2
1
A1F12
+Eb2 − J2
1− ε1
ε1A1
+J3 − J2
1
A2F23
= 0 (6)
at J3 ⇒ Q13 + Q23 + Q3 = 0
J1 − J3
1
A1F13
+J2 − J3
1
A2F23
+Eb3 − J3
1− ε1
ε1A1
= 0 (7)
12
if surface temperature is known
• given Ti, evaluate Ebi = σT 4i
• evaluate all space and surface resistances
• solve for J1, J2 and J3
• determine the heat flow rate as
Qi = Ai
∑Fij(Ji − Jj)
if surface heat flow rate is known
• replace Q1, Q2 and/or Q3 in Eqs. 1-3
• solve for J1, J2 and J3
• determine the surface temperature as
σT 4i = Ji +
1− εiεi
∑Fij(Ji − Jj)
Special Cases
The system of equations for 2 and 3-surface enclosures can simplify further when one or moresurfaces are: i) blackbody surfaces or ii) reradiating (fully insulated) surfaces.
blackbody surface: for a blackbody surface ε = 1 and the surface resistance goes to zero. As aconsequence the radiosity can be calculated directly as a function of surface temperature
Ji = Ebi = σT 4i
reradiating surface: Qi = 0, therefore the heat flow into the radiosity node equals the heat flowout of the node.
The resistor network simplifies to:
QRR
R
RR
Q Q
E Eb1 b221
12
13
23
212
13 23
1
13
and the system of equations can be easily solved as:
Q1 = −Q2 =Eb1 − Eb2
R1 +
[1
R12
+1
R13 +R23
]−1
︸ ︷︷ ︸parallel resistance
+R2
Example 7-1: Consider two very large parallel plates with diffuse, gray surfaces, Determinethe net rate of radiation heat transfer per unit surface area, Q12/A, between the two surfaces.For Case 2, also determine T3, the temperature of a radiation shield, positioned midway be-tween surfaces 1 and 2.
Given:
ε1 = 0.2 T1 = 800 Kε2 = 0.7 T2 = 500 Kε3 = 0.02 A1 = A2 = A3 = A
Assume steady state conditions.
14
Example 7-2: A thermocouple is suspended between two parallel surfaces as shown in thefigure below. Find Tf , the temperature of the air stream by performing an energy balance onthe thermocouple.
Given:
Tw = 400 K Tth = 650 Kεth = 0.6 h = 80 W/(m2 ·K)
Assume steady state conditions.
15
Example 7-3: Consider a room that is 4 m long by 3 m wide with a floor-to-ceiling dis-tance of 2.5 m. The four walls of the room are well insulated, while the surface of the flooris maintained at a uniform temperature of 30 ◦C using an electric resistance heater. Heat lossoccurs through the ceiling, which has a surface temperature of 12◦C. All surfaces have anemissivity of 0.9.
a) determine the rate of heat loss, (W ), by radiation from the room.b) determine the temperature, (K), of the walls.
16